Vascular smooth muscle cell (SMC) de-differentiation with subsequent migration and proliferation into the subendothelial space is central to the progression of cardiovascular diseases. The Nox4 NADPH oxidase (Nox4) is implicated in maintaining the differentiated phenotype of SMC in part through myocardin, a master regulator of SMC gene expression. However, this process is poorly understood. We hypothesized that microRNAs (miR)-mediate changes in Nox4 expression and regulate SMC differentiation. Treatment of human SMCs with a miR-9 or miR-25 mimic silenced Nox4 mRNA through binding to the Nox4 3’UTR. However, only miR-25 was sufficient to downregulate Nox4 protein levels. We found that miR-25 induced the expression of miR-9 through a novel mechanism involving demethylation of the miR-9 promoter by Tet methylcytosine dioxygenase 2 (TET2). Inhibition of miR-9 induction by miR-25 with a miR-9 inhibitor restored Nox4 protein expression to basal levels. Furthermore, the miR-25-mediated decrease in Nox4 protein was ameliorated by inhibiting the proteasome with MG132. These data suggest a novel mechanism wherein miR-9 and miR-25 regulate Nox4 through both translational suppression and proteosomal degradation. Overexpression of miR-9 or miR-25 in human SMCs (1) suppressed myocardin mRNA and protein expression; (2) decreased expression of multiple SMC differentiation genes; and (3) was sufficient to induce cell migration. Thrombin and tumor necrosis factor increased the expression of miR-9 and miR-25 in human SMCs and inhibition of miR-9 prevented thrombin-mediated decrease in myocardin and SMC migration. Mir-9 and miR-25 levels were increased in SMCs derived from balloon injured rat aorta as compared to medial SMCs and in murine carotid artery ten days post carotid injury. A miR-9 inhibitor decreased neointimal formation by more than 50% in following partial carotid ligation in mice. These findings identify miR-9/Nox4 as a novel regulatory pathway of SMC differentiation and a potential therapeutic target in vascular disease.

Obesity is an independent risk factor for cardiovascular disease, including atherosclerosis and restenosis. Obesity has been associated with low levels of chronic inflammation. We hypothesized that in chronic inflammatory disease states, like obesity, circulating epidermal growth factor (EGF)-like ligands accelerate vascular disease by upregulating the Nox1 NADPH oxidase expression and activity. Ten week old male and female mice were fed a 16% fat (Teklad, TD.94059, n=7) or normal chow (lean, n=9) diet for 15 weeks (weight 41.6 vs. 27.9 gms). Heparin bound (HB)-EGF levels were increased 2.9-fold in serum from obese mice as compared to lean litter mates. Thirty days after carotid ligation, neointimal formation was greater in the obese mice (intima/media area ratio 0.65 vs 0.25, obese vs. lean, p<0.05) and the systemic administration of AG1478, an EGFR inhibitor, by osmotic minipump, reduced neointimal formation in the obese mice (I/M 0.16, n=8). Lean mice exposed to increased circulationg HB-EGF by an osmotic minipump developed increased vascular expression of Nox1 (1.7-fold), but not Nox4. In cultured smooth muscle cells (SMC), HB-EGF increased Nox1 expression, NFkB activation, and cell proliferation (2.6-fold). The HB-EGF-mediated increase in cell growth was prevented by knockdown of Nox1 and by the mitochondrial targeted antioxidant, mitoTEMPO. Overexpression of Nox1 “rescued” the inhibitory effects of mitoTEMPO on SMC proliferation. These data suggest that obesity is associated with increased circulating EGF-ligands which contribute to the development of cardiovascular disease by a mechanism that involves mitochondrial ROS-induced ROS and the increased expression of Nox1.

Inhibition of vascular smooth muscle cell (VSMC) proliferation by drug eluting stents has markedly reduced intimal hyperplasia and subsequent in-stent restenosis. However, the effects of antiproliferative drugs on endothelial cells (EC) contribute to delayed re-endothelialization and late stent thrombosis. Cell-targeted therapies to inhibit VSMC remodeling while maintaining EC health are necessary to allow vascular healing while preventing restenosis. We describe an RNA aptamer (Apt 14) that functions as a smart drug by preferentially targeting VSMCs as compared to ECs and other myocytes. Furthermore, Apt 14 inhibits phosphatidylinositol 3-kinase/protein kinase-B (PI3K/Akt) and VSMC migration in response to multiple agonists by a mechanism that involves inhibition of platelet-derived growth factor receptor (PDGFR)-beta phosphorylation. In a murine model of carotid injury, treatment of vessels with Apt 14 reduces neointimal formation to levels similar to those observed with paclitaxel. Importantly, we confirm that Apt 14 cross-reacts with rodent and human VSMCs, exhibits a half-life of similar to 300 hours in human serum, and does not elicit immune activation of human peripheral blood mononuclear cells. We describe a VSMC-targeted RNA aptamer that blocks cell migration and inhibits intimal formation. These findings provide the foundation for the translation of cell-targeted RNA therapeutics to vascular disease.

Cardiovascular disease (CVD) is the leading cause of mortality in many countries. Many vascular disorders, including in-stent restenosis, arteriosclerosis, vein graft disease, and cardiac allograft arteriopathy are caused by pathological vascular smooth muscle cell (VSMC) remodeling following injury. An ideal therapeutic intervention would target the VSMCs without impairing the injured vessel re-endothelialization. However, current therapies do not selectively prevent pathological VSMC remodeling leading to impaired re-endothelization, late stent thrombosis and death. Thus, there is a clear need for cell-targeted treatment and prevention options of pathological VSMC remodeling. Our group has described the development of VSMC-specific, aptamers for (1) modulating signaling pathways associated with pathological VSMC remodeling and (2) delivering therapeutic molecules to these cells in vivo. Here we demonstrate that one of these aptamers, Vapt14, inhibits protein kinase B (PKB)/Akt activation and VSMC migration in response to multiple agonists by a mechanism that involves inhibition of platelet-derived growth factor receptor (PDGFR)-beta phosphorylation. In a murine model of carotid injury, treatment of vessels with Vapt14 reduces intimal:medial thickness to levels comparable to that of paclitaxel. Importantly, we confirm that Vapt14 cross-reacts with rodent and human VSMCs, exhibits a half-life of ~300 hours in human serum, and does not elicit immune activation of human peripheral blood mononuclear cells (PBMCs) in vitro. In addition, we confirm delivery of Vapt14 to VSMC in vitro and in vivo with fluorescence microscopy. Studies are being expanded to evaluate aptamer-mediated delivery of therapeutic biomolecules (e.g. small molecules, RNAi modulators) to areas of vascular injury. In summary this work provides an essential foundation for the translation of cell-targeted RNA therapeutics to multiple hyperplastic vascular diseases.

Chronic inflammation is associated with extracellular oxidative stress and accelerated vascular disease. We have reported that a more oxidized extracellular redox state promotes epidermal growth factor (EGF)-like ligand shedding, resulting in EGFR-mediated increase in expression of Nox1 NADPH oxidase and proliferation of smooth muscle cells (SMC). Furthermore, we observed increased plasma levels of EGF-like ligands, particularly HB-EGF, in a non-human primate model of atherosclerosis. We hypothesized that EGF-like ligands contribute to the development of vascular disease by promoting Nox1-mediated SMC growth. In cultured SMCs, treatment with HB-EGF increased Nox1 transcription and SMC proliferation. Expression of siRNA to Nox1, but not Nox4, prevented HB-EGF-induced cell growth. The mitochondria-targeted antioxidant MitoTEMPO inhibited HB-EGF-mediated Nox1 expression and SMC proliferation. Furthermore, overexpression of Nox1 restored HB-EGF-mediated proliferation in the presence of MitoTEMPO. We next implanted osmotic minipumps containing HB-EGF into 16-wk old hypercholesterolemic mice and 2d later performed carotid ligation as a model of vascular injury. Two weeks following ligation, mice receiving HB-EGF demonstrated a 60% increase in plasma HB-EGF, 65% increase in aortic expression of Nox1, and a 55% increase in carotid intimal area (I/M ratio 0.52±0.01 vs. 0.91±0.01, p 50%) were divided into two groups based on the severity of allograft vasculopathy as determined by angiography. HB-EGF levels positively correlated with disease severity, with circulating concentrations 1.9-fold higher in patients with moderate to severe coronary disease as compared to patients with absent to mild disease (n=9/group; p

Arterial revascularization by stenting is associated with restenosis due to proliferation of vascular smooth muscle cells (VSMCs) and intimal hyperplasia. Although the use of cell cycle inhibitors in drug eluting stents (DES) has reduced the rate of in-stent restenosis, this approach also inhibits re-endothelialization, thereby requiring prolonged antithrombotic regimens to prevent stent thrombosis. Delivery of nucleic acid aptamers by DES is a potential novel approach due to their high specific binding affinity and potential for modification by medicinal chemistry. The goal of this study was to identify RNA aptamers that (1) internalize into VSMCs and/or (2) specifically inhibit VSMC activation. Using an in vivo compatible RNA aptamer library, VSMC-specific aptamers were selected using a cell-internalization SELEX (systematic evolution of ligands by exponential enrichment) process with iterative rounds of positive selection using VSMCs and negative selection using endothelial cells (ECs). The top candidate aptamers demonstrated 15-20-fold specificity for internalization into cultured VSMCs vs . ECs. Specificity of these aptamers was confirmed by ex vivo studies that demonstrated preferential internalization into denuded, but not endothelium-intact, aortic segments. In addition, several of the VSMC-targeting aptamers exhibited differential effectiveness in inhibiting VSMC migration as measured by Boyden chamber assays. In proof-of-concept delivery studies, we next engineered aptamer-siRNA chimeras using siRNAs targeting Nox1 NADPH oxidase, which has been implicated in intimal hyperplasia. These VSMC-targeting aptamer-siRNA chimeras retain specificity for VSMCs and are being evaluated for efficacy in reducing intimal formation in a murine model of vascular injury. In conclusion, we developed novel VSMC-targeting aptamers that may serve as a platform technology to selectively deliver therapeutics to VSMCs over ECs. The aptamers that inhibited cell migration present an opportunity for dual-action VSMC-targeted therapeutic agents. Future refinement of this panel of VSMC-directed aptamers will enable broader application of targeted therapies for vascular-proliferative diseases.

AIMS:In atherosclerosis and restenosis, vascular smooth muscle cells (SMCs) migrate into the subendothelial space and proliferate, contributing to neointimal formation. The goal of this study was to define the signalling pathway by which Nox1 NAPDH oxidase mediates SMC migration.METHODS AND RESULTS:SMCs were cultured from thoracic aorta from Nox1(-/y) (Nox1 knockout, KO) and wild-type (WT) mice. In response to thrombin, WT but not Nox1 KO SMCs generated increased levels of reactive oxygen species (ROS). Deficiency of Nox1 prevented thrombin-induced phosphorylation of Src and the subsequent transactivation of the epidermal growth factor receptor (EGFR) at multiple tyrosine residues. Next, activation of extracellular signal-regulated kinase 1/2 (ERK1/2) and matrix metalloproteinase-9 (MMP-9) by thrombin was inhibited by the EGFR inhibitor AG1478 and in Nox1 KO SMCs. Thrombin-induced shedding of N-cadherin from the plasma membrane was dependent on the presence of Nox1 and was blocked by AG1478 and an inhibitor of metalloproteinases. Migration of SMCs to thrombin was impaired in the Nox1 KO SMCs and was restored by expression of Nox1. Finally, treatment of WT SMCs with AG1478 abrogated Nox1-dependent SMC migration.CONCLUSIONS:The Nox1 NADPH oxidase signals through EGFR to activate MMP-9 and promote the shedding of N-cadherin, thereby contributing to SMC migration.

Background/aims: Reduced activity of the antioxidant glutathione peroxidase-1 (GPx1) correlates with increased risk of cardiovascular events in patients with coronary artery disease. However, it remains unclear whether this imbalance in antioxidant capacity directly contributes to activation of vascular cells. In response to oxidative stress, smooth muscle cells (SMCs) secrete the pro-inflammatory immunomodulator cyclophilin A (CyPA). We hypothesized that reduction in vascular cell GPx1 activity causes secretion of CyPA and paracrine-mediated activation of NF-kappa B and proliferation of SMCs.Methods/results: Using a murine model of GPx1 deficiency (GPx1(+/-)), we found elevated levels of hydrogen peroxide levels and increased secretion of CyPA in both arterial segments and cultured SMCs as compared to wild type (WT). Conditioned media from GPx1(+/-) SMCs caused increased NF-kappa B activation of quiescent WT SMCs, and this was inhibited by the antioxidant N-acetyl-L-cysteine or by cyclosporine A (CsA). In co-culture experiments. SMCs derived from GPx1(+/-) aorta caused increased proliferation of WT SMCs, which was also inhibited by CsA.Conclusions: Reduction in vascular cell GPx1 activity and the associated increase in oxidative stress cause CyPA-mediated paracrine activation of SMCs. These findings identify a novel mechanism by which an imbalance in antioxidant capacity may contribute to vascular disease. Published by Elsevier Inc.

Redox-dependent migration and proliferation of vascular smooth muscle cells (SMCs) are central events in the development of vascular proliferative diseases; however, the underlying intracellular signaling mechanisms are not fully understood. We tested the hypothesis that activation of Nox1 NADPH oxidase modulates intracellular calcium ([Ca(2+)](i)) levels. Using cultured SMCs from wild-type and Nox1 null mice, we confirmed that thrombin-dependent generation of reactive oxygen species requires Nox1. Thrombin rapidly increased [Ca(2+)](i), as measured by fura-2 fluorescence ratio imaging, in wild-type but not Nox1 null SMCs. The increase in [Ca(2+)](i) in wild-type SMCs was inhibited by antisense to Nox1 and restored by expression of Nox1 in Nox1 null SMCs. Investigation into potential mechanisms by which Nox1 modulates [Ca(2+)](i) showed that thrombin-induced inositol triphosphate generation and thapsigargin-induced intracellular calcium mobilization were similar in wild-type and Nox1 null SMCs. To examine the effects of Nox1 on Ca(2+) entry, cells were either bathed in Ca(2+)-free medium or exposed to dihydropyridines to block L-type Ca(2+) channel activity. Treatment with nifedipine or removal of extracellular Ca(2+) reduced the thrombin-mediated increase of [Ca(2+)](i) in wild-type SMCs, whereas the response in Nox1 null SMCs was unchanged. Sodium vanadate, an inhibitor of protein tyrosine phosphatases, restored the thrombin-induced increase of [Ca(2+)](i) in Nox1 null SMCs. Migration of SMCs was impaired with deficiency of Nox1 and restored with expression of Nox1 or the addition of sodium vanadate. In summary, we conclude that Nox1 NADPH oxidase modulates Ca(2+) mobilization in SMCs, in part through regulation of Ca(2+) influx, to thereby promote cell migration.

Objective-We have shown that the chloride-proton antiporter chloride channel-3 (ClC-3) is required for endosome-dependent signaling by the Nox1 NADPH oxidase in SMCs. In this study, we tested the hypothesis that ClC-3 is necessary for proliferation of smooth muscle cells (SMCs) and contributes to neointimal hyperplasia following vascular injury.Methods and Results-Studies were performed in SMCs isolated from the aorta of ClC-3-null and littermate control (wild-type [WT]) mice. Thrombin and tumor necrosis factor-alpha (TNF-alpha) each caused activation of both mitogen activated protein kinase extracellular signal-regulated kinases 1 and 2 and the matrix-degrading enzyme matrix metalloproteinase-9 and cell proliferation of WT SMCs. Whereas responses to thrombin were preserved in ClC-3-null SMCs, the responses to TNF-alpha were markedly impaired. These defects normalized following gene transfer of ClC-3. Carotid injury increased vascular ClC-3 expression, and compared with WT mice, ClC-3-null mice exhibited a reduction in neointimal area of the carotid artery 28 days after injury.Conclusion-ClC-3 is necessary for the activation of SMCs by TNF-alpha but not thrombin. Deficiency of ClC-3 markedly reduces neointimal hyperplasia following vascular injury. In view of our previous findings, this observation is consistent with a role for ClC-3 in endosomal Nox1-dependent signaling. These findings identify ClC-3 as a novel target for the prevention of inflammatory and proliferative vascular diseases. (Arterioscler Thromb Vasc Biol. 2011;31:345-351.)

The transmembrane glycoprotein N‐cadherin (N‐cad) has a critical role in the dynamic rearrangement of cell to cell adhesion. Reduction in the level of cell surface N‐cad by proteolytic shedding increases transcriptional activity of genes involved in cell migration. Although NADPH oxidases participate in cell growth, their role in shedding of N‐cad is unknown. We hypothesized that the NADPH oxidase subunit Nox1 contributes to shedding of N‐cad and migration of smooth muscle cells (SMC). Studies were performed in SMC cultured from wildtype (WT) and Nox1 null mice. Treatment of WT SMC with thrombin reduced membrane levels of N‐cad, which was prevented by the protease inhibitor MG132, suggesting N‐cad shedding. Pretreatment with antisense to Nox1 abolished the thrombin‐mediated shedding of N‐cad. Similarly, thrombin failed to reduce N‐cad levels in Nox1 null SMC. In WT SMC, cleavage of N‐cad and activation of matrix‐metalloproteinase (MMP)‐9 by thrombin was dependent on transactivation of EGFR. Activation of EGFR and MMP‐9 were significantly reduced in Nox1 null SMC. Finally, WT SMC migration was attenuated by the EGFR inhibitor AG1478 or by Nox1 antisense. These data suggest that Nox1‐mediated EGFR transactivation promotes proteolysis of N‐cad, potentially by MMP‐9, disrupting cell‐cell adhesions important in SMC migration.

One of alcohol's most damaging effects on the developing CNS is severe neuronal loss. For unknown reasons, the cerebellum and hippocampus are more vulnerable to alcohol‐induced neuronal loss than other regions. One possibility is that alcohol‐resistant regions have protective mechanism(s), which mitigate alcohol toxicity. Our prior studies revealed that NO is protective, reducing alcohol‐induced neuronal loss by activating a NO‐signaling pathway (NO activates guanylyl cyclase; the increase in cGMP activates cGMP‐dependent protein kinase, PKG). A current goal is to identify a downstream target for PKG. We are exploring interactions among PKG, alcohol‐induced neuronal death, and [Ca2+]¡ for several reasons. Alcohol disrupts [Ca2+]¡ homeostasis, and neurons are sensitive to calcium disruption. PKG regulates the inositol 1,4,5‐triphosphate receptor (IP3R), modulating [Ca2+]¡ release. Ethanol (400 mg/dl) has two effects on CGN cultures: 1) A 25% reduction in neuronal numbers; 2) A rapid (< one min) increase in [Ca2+]¡. Calcium chelators (BAPTA‐AM; EGTA‐AM) decrease alcohol‐induced neuronal loss (<10%). An inhibitor of the IP3R (2‐APB) and a PKG activator (8‐Br‐cGMP) reduce both the alcohol‐induced [Ca2+]¡ surge and neuronal death (<5%). In summary, modulating [Ca2+]¡ reduces alcohol toxicity. NO may protect neurons by preventing alcohol's disruption of [Ca2+]¡. Support: NIAAA grant AA011577

Mechanisms by which NADPH oxidase‐derived ROS modulate cell signaling pathways are poorly understood. Using SMC cultured from aorta of Nox1 null and littermate control (WT) mice, we tested the hypothesis that ROS generated by Nox1, a catalytic subunit of NADPH oxidase, modulate thrombin‐induced increases in intracellular calcium ([Ca2+]I). In WT SMC, thrombin (1 U/ml) rapidly increased [Ca2+]I with a peak of 1100±70 nM Ca2+ (n=89 cells), as measured by fura‐2 fluorescence ratio imaging, which was inhibited by antisense to Nox1. Similarly, thrombin‐induced increase in [Ca2+]I was markedly attenuated in Nox1 null SMC (620±70 nM Ca2+, n=114 cells, p<0.05). Investigation into potential mechanisms by which Nox1‐derived ROS modulate [Ca2+]I showed that thrombin‐induced generation of inositol trisphosphate and the release of intracellular Ca2+by thapsigargin were similar in WT and Nox1 null SMC. In contrast, bathing cells in Ca2+‐free media significantly reduced the thrombin‐mediated increase of [Ca2+]I in WT SMC, whereas the response in Nox1 null SMC was unchanged. Sodium vanadate, an inhibitor of protein tyrosine phosphatases (PTPs), normalized the thrombin‐induced increase of [Ca2+]I in Nox1 null SMC. These data suggest that inhibition of PTPs by Nox1‐derived ROS mediate influx of extracellular Ca2+ in SMC, and identify a novel signaling mechanism that may have a role in vascular disease.

NADPH oxidase-derived reactive oxygen species (ROS) contribute to the pathobiology of vascular disease. However, studies investigating NADPH oxidases in atherosclerosis have been limited in their ability to distinguish between the role of vascular cell and inflammatory cell -derived ROS. In this study, we examined the contribution of Nox1-derived ROS in a mouse model of atherosclerosis. Nox1 is a primary catalytic subunit of vascular cell, but not inflammatory cell, NADPH oxidase. At weaning, male apolipoprotein E deficient mice (AS, n=12) and male mice deficient in both apolipoprotein E and Nox1 (AS/Nox, n=16) received an atherogenic diet for 18 weeks. Mean blood pressures (116±3 vs. 110 ±3 mmHg; AS vs. AS/Nox, n=6), weights, and serum cholesterol levels (1578±150 vs. 1631±80 mg/dl; AS vs. AS/Nox) were similar between the AS and AS/Nox mice. As measured by lucigenin-enhanced chemiluminescence, superoxide levels were increased in segments of thoracic aorta from AS mice as compared to aorta from control mice (22±2 vs. 14±2 RLU/sec/mm 2 ; AS vs. C57BL/6; p<0.05, n=6). In contrast, superoxide levels in segments of thoracic aorta from AS/Nox mice were significantly lower (9±1 RLU/sec/mm 2 , n=6) than both AS and C57BL/6 mice. Dihydroethidium staining confirmed decreased superoxide levels in aorta of AS/Nox mice. Atherosclerotic lesion area was measured by staining the aorta en face with Oil Red O. Although atherosclerotic lesion area was reduced over the entire length of aorta in AS/Nox mice as compared to AS mice (10±1% vs. 14±1%, p<0.01), the reduction in lesion was primarily limited to the aortic arch (28±3% vs. 43±2%, p<0.001). In summary, Nox1 contributes to generation of ROS and lesion formation in atherosclerosis. These data confirm a role for vascular cell NADPH oxidases, and in particular Nox1, in vascular disease.